The Q/U Imaging ExperimenT (QUIET) receivers Coherent Polarimeter Arrays at 40 and 90 GHz

Transcription

1 The Q/U Imaging ExperimenT (QUIET) receivers Coherent Polarimeter Arrays at 40 and 90 GHz Dorothea Samtleben, Max-Planck-Institut für Radioastronomie, Bonn

2 Universe becomes transparent => Release of Cosmic Microwave background Radiation

3 Density fluctuations produce E-modes, B-modes derive from Lensing and Primordial Gravity Waves Size of B-modes from Primordial Gravity Waves still unknown, Parametrized by Tensor-to-Scalár Ratior Chiang et al,

4 Picture of the Field A Program of Technology Development and of Sub-Orbital Observations of the Cosmic Microwave Background Polarization Leading to and Including a Satellite Mission ASTRO 2010 Decadal Survey White Paper

5 The QUIET Collaboration 14 institutes, 5 countries, ~30 people Caltech, Chicago, Columbia, Fermilab, JPL, KEK (Japan), Manchester, Miami, (Michigan), MPIfR, Oslo, Oxford, Princeton, Stanford

6 Radiometer on a chip: QUIET Receiver Module InP MMIC GaAs Schottky Diodes Automated assembly and optimization Large array of correlation polarimeters Measuring Q/U simultaneously in each pixel Complementing frequencies from other experiments (bolometric experiments are >100 GHz) ~1 inch Produced by JPL based on developments for Planck LFI

7 QUIET L/R Correlator: Simultaneous Q/U measurements E y E a E b LNA LNA LNA LNA E x ±1 ±1 4kHz phase switching Q U

8 Differential Total Power Receivers (MPIfR) Design and production of differential total power receivers: 2 Q-band, 6 W-band Measuring T between neighboured beams, phase switching reduces 1/f of amplifiers Identification/characterization of unpolarized foregrounds Measurements of CMB Temperature and Temperature-Polarization correlations

9 Differential Total Power Receivers (MPIfR) OMT Feed 1 Ex1 Magic Tee Ey1 Feed 2 Ey2 Ex2 Magic Tee LNA LNA LNA LNA LNA LNA LNA LNA ±1 ±1 ±1 ±1 180 coupler 180 coupler => Tx1 - Ty2 Temperature difference Sensitive to polarization Module

10 Differential Total Power Receivers (MPIfR) W-band Q-band 2 of the JPL modules for the TT assemblies were assembled and tested at MPIfR with Frank Schäfer, Sener Türk

11 (Double) Demodulation Sampling at 800 khz Switching at 4 khz Blanking of 10% Combining to 100 Hz Large phase switch imbalance (mainly W-band) Slightly different frequency dependences of the diode responsivities => not all diodes simultaneously balanced => Use new demodulation scheme which eliminates phase switch imbalances: Leg A: kHz Leg B: Hz

12 50 Hz Time streams (in Chile, azimuthal scan, constant elevation) unswitched FFT Switched RMS=0.05mV

13 Window Receivers for QUIET Horn array Build large receiver arrays in cryostats Install up to 3 telescopes (1.4m) in the Atacama Desert Septum Polarizers 84+6 * pixel 90 GHz FWHM 12 array sensitivity: 55 µk s Phase I, in Chile 2008/09 Modules, in dewar electronics 17+2 * pixel 40 GHz FWHM 28 array sensitivity: 65 µk s * 6 (2) pixels are Total Power Pixels in the W (Q) band array Cryostat 3 x 499 pixel 90 GHz 61 pixel 40 GHz 18 pixel 30 GHz Phase II

14 Q-band array (40 GHz) ~1 inch Horn arrays from 100 platelets combined by diffusion bonding

15 W-band pictures! W-band array (90 GHz)

16

17 Characterization Assembly and Integration of the arrays in Columbia (Q-band) and Chicago (W-band) Two-load tests (liquid Nitrogen, Argon, Oxygen and 300K eccosorb) Band Sweeps Gain determination and optimization with rotating metal plate reflected from cold load

18 Wire spacing: 0.5inch Automatic optimization Per module10 gate/drain voltages starting with JPL values Downhill-simplex optimization iterations (coarse) ~50 iterations (fine) Whole array optimized in 10 hours Sparse wire grids with inch spacing

19 Focal Plane (Receiver) Absorptive ground screen Platelet Array 2nd Mirror Electronics Box Primary Mirror Mount

20 Observing site: Chajnantor Plateau in the Atacama Desert in Chile, 5000m altitude Extremely dry site Observing year round QUIET

21 Rotating wiregrid at the site looking at the sky (via metal plate)

22 Rotating wiregrid at the site looking at the sky (via metal plate)

23 Histogram of data/white noise error for one channel PRELIMINARY Histogram of widths over the season 70% observing efficiency 90% of observing time for CMB data

24 Calibration Jupiter in Total Power Receivers Tau A in Polarization Temperature: - Elevation nods (10% error) - Jupiter/Venus/RCW38 (5% error) Elevation nod in unswitched Total Power channels Polarization: - Tau A (10% error) - Elevation nods (I->Qleakage to 0.1%) - Moon (angle uncertainty 2 ) - Noise Source (<5% error) - Wire Grid (1% error on relative gain) Moon in Polarization

25 Preliminary sensitivities/element Performance/Improvements Q-band W-band Issues: ~5% channels dead/unusable 1-2% Leakage (septum polarizer) Small bandwidth (hybrid) Higher noise temperatures than expected from LNAs Q / W mk µk s Sensitivity/element 0.27 / 0.5 /array 65 / 55 Noise temperatures: 30 / 90 K Bandwidth: 8 / 12 GHz 1/f knees: 20 / 20 mhz s Improvements for phase II: Automatic assembly (automatically assembled modules needed less rework) Selection of chips from cryogenic testing (before warm selection) Changing from 100nm to 35nm gates (for W-band noise temperatures of ~30K were measured) Adapt W-band hybrid design to Q-band design for larger bandwidth Exchange detector diodes with ones that require no bias when cold

26 Galactic center map from Total Power Receivers (2 Q band receivers)

27 CMB Field (Total Power) QUIET/ WMAP

28 Galactic center map from Polarization Receivers (17 Q band receivers) QUIET simulation WMAP

29 Outlook Production of 3x499 W-band elements 61 Q-band elements 18 Ka-band elements Module mass production at Fermilab (W-band) and Stanford (Q/Ka-band) Cryostat window diameter 42 inch 4 coldheads/cryostat